Igal @ Lucee.org wrote:
> On 1/19/2016 10:58 AM, Alvaro Herrera wrote:
> >Yes, probably something like that. I think it failed the first time
> >because there was a bug (the one I introduced in a967613911f7), then
> >probably changed to src/backend and ran compiles there which probably
> >worked fine, leading to commit fa838b555f90. I might or might not have
> >removed the complete build dir instead of "make distclean"; not sure TBH.
> >As I recall, I tried a couple of builds after I pushed the fix commit and
> >couldn't get them to run at all. But since I saw in buildfarm that other
> >mingw builds were working, I lost interest. Then I deleted the mingw build
> >tree and didn't get back to retrying. TBH the only reason I mingled with
> >mingw at all is that nobody seems interested in fixing Win32 issues, and
> >this patch had been lingering for far too long.
>
> Ok, so your comments here made me think that maybe there is a bug somewhere
> in the distribution that I was using ( from
> http://www.postgresql.org/ftp/source/v9.5.0/ ), so I went ahead and cloned
> the git repo and used the sources from there.
>
> I am happy to announce that I was finally able to build and run PostgreSQL
> on MinGW-w64, but it only worked with the sources from the git repo, and not
> with the official downloads from the links above. I'm not sure if there is
> a bug there that was resolved, or if it is packaged differently, but I was
> using the official download because I read in several places that it should
> be easier to build from that distribution as opposed to the git repo.
Hmm, I guess this makes sense. A tree obtained from a shipped tarball
has "make distprep" already run, which means it contains a bunch more
files than clean Git workspace -- probably errcodes.h among them, for
example. This smells like a makefile bug of some kind to me.
--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
